Embedded / RT programming
Discussion
I'm coming at this with zero knowledge of the area and hoping that a few people within the PH font of all knowledge might be able to give me a few pointers / urls / books to read up.
Basically, I'm planning on building myself an oven for curing pre-preg composites under vacuum consolidation.
Me being me I don't want to simply use an on/off switch, because that would be far too easy
What I actually have in mind is a set of thermocouples at various locations within the oven, an embedded PC running Linux and controlling heater elements / fans, etc to allow precise curing times, temperatures, etc.
Can anyone here point me in the right direction? Is there an idiots guide to embedded Linux for system control?
Basically, I'm planning on building myself an oven for curing pre-preg composites under vacuum consolidation.
Me being me I don't want to simply use an on/off switch, because that would be far too easy
What I actually have in mind is a set of thermocouples at various locations within the oven, an embedded PC running Linux and controlling heater elements / fans, etc to allow precise curing times, temperatures, etc. Can anyone here point me in the right direction? Is there an idiots guide to embedded Linux for system control?
You probably don't need to use linux to do it. You might manage to use a Basic Stamp II if the algorithms aren't too complex.
Tim
Tim
The posh approach is to buy a computer which is designed to be used as an embedded controller, you will need the development tools to go with it. There are plenty to choose from. The cheap and cheerful approach is to buy a small PC. You can use whatever OS you want but for example a simple DOS application could be booted straight off a floppy disk and would run happily on a door stop spec PC.
Thanks for the input guys.
Tim, I think the BASIC Stamp might be a little limited for what I want, but their Javelin looks intersting - especially as Java is my day job.
And as GreenV8s said, I could just use an old PC. So the next step is trying to find a book or tutorial on writing the software and building the circuits to support this. Any ideas anyone?
Tim, I think the BASIC Stamp might be a little limited for what I want, but their Javelin looks intersting - especially as Java is my day job.
And as GreenV8s said, I could just use an old PC. So the next step is trying to find a book or tutorial on writing the software and building the circuits to support this. Any ideas anyone?
LexSport said:
Thanks for the input guys.
Tim, I think the BASIC Stamp might be a little limited for what I want, but their Javelin looks intersting - especially as Java is my day job.
And as GreenV8s said, I could just use an old PC. So the next step is trying to find a book or tutorial on writing the software and building the circuits to support this. Any ideas anyone?
Java programmer eh? Reet then - why not look into using a PIC Microcontroller - they are dirt cheap (like a few quid each) you program them in C which is easy enough if you've done Java.
You'll need a programmer but there are heaps of circuit layouts on the web.
There are some pretty powerfull PICs out there now, most have analogue inputs that you could hookup to a thermocouple and write a few lines of code to turn on/off fans/heaters/ & whatever.
Stick 'PIC microcontroller projects' into google and see wher eit gets you.
best
Ex
I've heard of PIC controllers, but at that point my knowledge of them ceases.
Google here I come.
As for the devices, it'll be mainly as above, thermocouples at various locations for heat sensing, heater elements to heat the thing up, fans to distribute the heat.
The logic will be to control the increase in temperature to a gradual 1*/min or so ramp then hold temp for x minutes, so I guess some way of selecting and displaying these variables.
This is why initially I was thinking along the lines of embedded Linux so I could link it up to an LCD display - I think having played with an Autoclave I'm possibly being a little ambitious...
Google here I come. As for the devices, it'll be mainly as above, thermocouples at various locations for heat sensing, heater elements to heat the thing up, fans to distribute the heat.
The logic will be to control the increase in temperature to a gradual 1*/min or so ramp then hold temp for x minutes, so I guess some way of selecting and displaying these variables.
This is why initially I was thinking along the lines of embedded Linux so I could link it up to an LCD display - I think having played with an Autoclave I'm possibly being a little ambitious...

LexSport said:
I'm possibly being a little ambitious...
Not at all...
Hey Darran,
Seriously, PICs are pretty easy to program - if I remember correctly people have even written multithreaded operating systems for them now!
With a bit of google time you'll be able to gather together all the ancillary component circuitry you need to run your idea.
I got into PICs as I wanted to make a GPS based datalogger for the rally car. I already had some GPS receivers that would drive a serial port.
I built a few extra bits and hooked up the PIC to the GPS and also logged the car RPM, gear, and brake light switch.
Admittedly I then streamed all this data out via another serial port to a PC104 CPU board (I havea few of these if you want one but I think they are overkill for your app) in order to save the data and then download it later to a laptop.
Its not 'rocket science', honest!
At the same time I also got the PIC to stream all this data out to an LCD display (via another serial port) so I could read it whilst driving. (LCD came from the US and had a blue screen - super sexy - I can try and dig out a URL if you want)
I'm not sure how good your electronic skills are but if you know the difference between a resistor and a capacitor you shouldn't have too much trouble once you set up the circuits to support the PIC (PSU, clock etc).
Get hold of some bread board and start playing/prototyping.
The first thing I started on was getting the PIC to flash an LED at a rate governed by a potentiometer.
Once you get your PIC programmer up and running and get some breadboard you can play to your heart's content.
After you've had a read around email me offline or post here if you want, I'll try and point you in the right direction with any questions you have.
best
Ex
Second that. PICs rock.
Agree re programming them in C. They're a bit short of test-and-branch instructions and have a weird two-bank register arrangement; it's good to have the C compiler look after all this for you. Do read the docs on the machine language though so you know what the underlying capabilities and limitations are.
The gadget for getting the code from the PC to the PIC is basically a few buffers hooked up to the printer port. I made my own and wrote the software to go with it. I could dig out the source code and the circuit diagram if you like.
Agree re programming them in C. They're a bit short of test-and-branch instructions and have a weird two-bank register arrangement; it's good to have the C compiler look after all this for you. Do read the docs on the machine language though so you know what the underlying capabilities and limitations are.
The gadget for getting the code from the PC to the PIC is basically a few buffers hooked up to the printer port. I made my own and wrote the software to go with it. I could dig out the source code and the circuit diagram if you like.
For ease of hardware interfacing and programming, buy a PLC from eBay.
You can purchase kit that originally cost thousands of pounds for just a few quid, and all the hardwork (actually interfacing to the outside world) has been done, and is reliable.
Depends what you want from your project though - we use Montavista Linux on some of our large scale control systems, with dedicated hardware modules, but for a single cell device like your oven, I'd pick a PLC any day.
In fact, a LOGO www.ad.siemens.de/logo/html_76/original.htm would do the trick, and they are very cheap.
You can purchase kit that originally cost thousands of pounds for just a few quid, and all the hardwork (actually interfacing to the outside world) has been done, and is reliable.
Depends what you want from your project though - we use Montavista Linux on some of our large scale control systems, with dedicated hardware modules, but for a single cell device like your oven, I'd pick a PLC any day.
In fact, a LOGO www.ad.siemens.de/logo/html_76/original.htm would do the trick, and they are very cheap.
Many thanks everyone for the info.
I knew I'd be able to count on PH to point me in the right direction
I can see there being a few pointless electronic devices being appearing around chez LexSport while I try to figure out what I'm doing with all this stuff.
Who knows, dual zone climate control in the Libra I'm building might not be too crazy an idea.
I knew I'd be able to count on PH to point me in the right direction
I can see there being a few pointless electronic devices being appearing around chez LexSport while I try to figure out what I'm doing with all this stuff.
Who knows, dual zone climate control in the Libra I'm building might not be too crazy an idea.

Logos with analog inputs on ebay for 5 quid:
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&category=4661&item=3831187994&rd=1
http://cgi.ebay.co.uk/ws/eBayISAPI.dll?ViewItem&category=4661&item=3831187994&rd=1
Gassing Station | Computers, Gadgets & Stuff | Top of Page | What's New | My Stuff



